90 Days After WMS Go-Live: The Health-Check Scorecard Your Implementer Won't Give You

Ninety days after your WMS go-live, the system isn't "new" anymore. But do you actually know whether it's healthy?

Go-live day, everybody's cheering: the consultants leave, the project manager gets promoted, the boss posts about it on LinkedIn. And then? Then nobody looks at the system reports for three months, until the day something blows up. A 3PL warehouse I work with learned this the hard way: in the run-up to peak season 2025, exactly 90 days after their new WMS went live, inventory accuracy had slid from 99.3% at launch to 97.6%. The books showed 400+ units of "ghost inventory" - in the system, not on the shelves. Peak-season replenishment was planned off the wrong numbers: bestsellers stocked out, slow movers piled up, and the whole thing ended with an all-night manual count.

My biggest takeaway: go-live isn't the finish line. Day 90 is the first real checkpoint. This post is the "90-day health-check scorecard" I've built up over the years watching systems and staring at data. Seller or warehouse owner, you can lift it and use it today.

90 Days After Go-Live: Which Metrics Actually Matter?

Unpopular opinion first: don't let the implementer's "the system is running fine" fool you. "Running" and "running well" are two completely different things. I track six numbers, each with a red line:

Those targets are marked "example" - every warehouse's SKUs and volumes are different, so don't copy them blindly. Use your own first month post-launch as the baseline, and at day 90 you're judging the trend, not the absolute value.

Counterintuitive: Rising UPH Can Be Bad News

Same warehouse. Month two after go-live, pick UPH was up 18%, and the ops manager was thrilled. Then I checked accuracy: the audit catch rate had spiked from 0.3% to 1.1%. Workers chasing UPH had "optimized" the verification step - scan verification downgraded to a quick glance.

UPH and accuracy are natural enemies; either one alone will lie to you. During the health check, read them together: UPH up and accuracy down means someone's cutting corners; both down means the process or the system genuinely has a problem; both up - that's when the system has truly gone live.

There's a sneakier trap: user activity. The implementer reports "120 daily active users, great engagement." Sounds nice, right? I once saw a warehouse where the three most "active" accounts belonged to the receiving team - because their receiving screen lagged and each order needed three logins. All that activity was forced. High login frequency doesn't mean the system works well; it might mean it's painful to use. Always pair activity with "average task completion time," or you're just cheering yourself on.

Red Flags That Mean the Implementation Never Actually Finished

Metrics are the lab report; these are the symptoms. Any one of them should worry you:

First, paper tickets are still circulating at day 90. Picks printed on paper, receiving logged by hand - the system workflow was never truly adopted. The implementation did a "launch," not a "landing."

Second, exception orders keep piling up. Exception handling is the easiest part of an implementation to fudge: "go live first, deal with exceptions later." When the exception pool passes 200 orders at day 90, "later" has effectively never arrived.

Third, nobody maintains master data. New SKUs take three days to set up, new location codes get passed around in spreadsheets - the system is running on dirty data, and pretty reports are just self-deception.

Fourth, only IT can pull reports. Business leads can't read the system dashboards and need IT's help every week - the training covered button-clicking, not "managing the warehouse with data."

How to Make This Scorecard Stick in Your Weekly Meeting

The scorecard exists; the question is how to use it. My approach is crude but works: 30 minutes every Monday morning, ops, IT, and customer service walk through it together. Each metric gets one of three states: green (on target), yellow (trending wrong), red (red line breached). Yellows get an owner and a deadline; reds need action that same week.

When day 90 arrives, hold a formal review: compare against your pre-launch baseline (old system or manual era) and do the full math - how much accuracy improved, how many heads saved, how much faster exceptions clear. That math serves two purposes: showing the boss the return, and showing the team the last three months weren't wasted.

After Day 90: Make the Health Check a Quarterly Habit

Don't file the scorecard away after day 90. Its best use is as a recurring ritual - run it every quarter, with baselines that move with the business.

Why quarterly? Because warehouse business is seasonal: Q4 peak-season UPH baselines and Q1 off-season baselines aren't even the same sport. Judge the off-season by peak standards and your people will quit; judge peak season by off-season standards and you won't see the blowup coming. At the start of each quarter, spend half a day resetting targets off last quarter's actuals. The red lines don't move, though - red lines are the floor, not the weather.

There's a bonus use: quarterly scorecards are your best leverage at renewal time. WMS annual support fees, the implementer's on-site contract - when they come up for renewal, four quarters of scorecards on the table beat any sales pitch. System performing well, the money was well spent; performing badly, that's your discount - or your reason to switch vendors. Day to day it's a management tool; at the negotiating table it's a weapon.

Blunt truth to close: the real value of a 90-day WMS health check isn't finding bugs. It's forcing you to answer one question - is this system "installed," or is it actually "used"? Installed is IT's job. Used is where the money is.

Tried and tested by me - take it, tweak the targets, and it's your warehouse's version.

By Ah K, E-commerce Seller (5 yrs)